Armed-disarmed effects in carbohydrate chemistry: History, synthetic and mechanistic studies

Abstract

This chapter begins with an account of the serendipitous events that led to the development of n-pentenyl glycosides (NPGs) as glycosyl donors, followed by the chance events that laid the foundation for the armed-disarmed strategy for oligosaccharide assembly. A key mechanistic issue for this strategy was that, although both armed and disarmed entities could function independently as glycosyl donors, when one was forced to compete with the other for one equivalent of a halonium ion, the disarmed partner was found to function as a glycosyl acceptor. The phenomenon was undoubtedly based on reactivity, but further insight came unexpectedly. Curiosity prompted an examination of how ω-alkenyl glycosides, other than n-pentenyl, would behave. Upon treatment with wet N-bromosuccinimide, allyl, butenyl, and hexenyl glucosides gave bromohydrins, whereas the pentenyl analog underwent oxidative hydrolysis to a hemiacetal. Although the answer was definitive, an in depth comparison of n-pentenyl and n-hexenyl glucosides was carried out which provided evidence in support of the transfer of cyclic bromonium ion between alkenes in a steady-state phenomenon. It was found that for two ω-alkenyl glycosides having a relative reactivity ratio of only 2.6:1, nondegenerate bromonium transfer enabled the faster reacting entity to be converted completely to product, while the slower reacting counterpart was recovered completely. This nuance suggests that in the armed/disarmed coupling, such a nondegenerate steady-state transfer is ultimately responsible for determining how the reactants are relegated to donor or acceptor roles. Development of chemoselective armed/disarmed coupling led to another phase in the sequence of serendipities. During experiments to glycosylate an acceptor diol, it was found that armed and disarmed donor's glycosylated different hydroxyl groups. This observation caused us to embark on studies of regioselective glycosylation. One of these studies showed that it is possible to activate selectively n-pentenyl orthoesters (NPOEs) over other n-pentenyl donors, and that this chemoselective process enables regioselective glycosylation. As a result, reaction partners can be so tuned that glycosylation of an acceptor with nine free hydroxyl groups by an n-pentenyl orthoester donor carrying two free hydroxyl groups is able to furnish a single product in 42% yield. Experiments such as the latter suggest that the donor favors a particular hydroxyl group, and/or that a particular hydroxyl favors the donor. Either option implies that the principle of reciprocal donor acceptor selectivity (RDAS) is in operation. Such examples of regioselective glycosylation provide an alternative to the traditional practice of multiple protection/deprotection events to ensure that the only free hydroxyl group among glycosyl partners is the one to be presented to the donor. By avoiding such protection/deprotections, there can be substantial savings of time and material - as well as nervous anxiety. © 2011 Springer-Verlag Berlin Heidelberg.